This application directly addresses three challenge areas: first broad Challenge Area (01) Behavior, Behavioral Change, and Prevention (specific topic 01-AG-102 Neural mechanisms), second broad Challenge area (03) Biomarker Discovery and Validation (specific topic 03-AG-111 bridge animal model) and third broad Challenge area (05) Comparative Effectiveness Research (specific topic "05-AG-105* Diseases and Syndromes of Aging Including Neurodegenerative Diseases"). We will evaluate the potential impact on biomarkers and cognition of a combined dietary and exercise paradigm (DE), using two different types of animal models for age-related Alzheimer's disease (AD) amyloid and tau pathology and related neurodegeneration. The experiments described will define how these treatments or combined treatments impact cognition, neuronal and synaptic proteins and alterations in CSF biomarkers. AD, the most common cause of dementia, is projected to rapidly deplete Medicare as our population ages, arguing for the need of an AD prevention program with minimal adverse side-effects and cost issues. Challenge 1) Multiple pathways appear to contribute to AD progression. Because AD pathogenesis is complex with different stages and pathologies occurring in different brain regions simultaneously, individual treatments may have to be combined for optimal results. Our data suggest that the DE components proposed synergize in multiple ways to limit both the [unreadable]-amyloid (plaque) and tau (tangle) pathology and associated oxidative damage and inflammation. Thus we propose to develop an optimized combined therapy program, which targets AD pathogenesis and may also limit other age-related pathologies. Challenge 2) Single models for AD are inadequate. Most proposed AD treatments target only A[unreadable] or tau and most have been applied prior to lesions in incomplete AD models, typically models only for A[unreadable]. Thus we propose to use dual models: one that evaluates [unreadable]-amyloid (A[unreadable]) dependent cognitive deterioration and one that evaluates tau- dependent cognitive deterioration and neurodegeneration. Challenge 3) Biomarkers in models should be identified that can predict clinical change in trials. Thus we will employ a transgenic rat model where we have demonstrated that neuroimaging and CSF biomarkers can be readily tracked (Aim 1). Our data suggest that the omega-3 fatty acid docosahexaenoic acid (DHA) and the phenolic antioxidant curcumin (from turmeric) synergize to reduce AD pathogenesis in both APP/ A[unreadable] and tau pathology models. In non-AD models, synergism between DHA and exercise has been demonstrated. Therefore we propose a combined therapy program to test dietary (DHA, curcumin) and exercise interventions. Because there are no ideal models for AD, we propose to use two models that isolate amyloid plaque (Aim 1, a transgenic APP/PS1 rat model) or tangle/ neuron loss (Aim 2, wildtype human tau model). Variables measured will be cognitive function, synaptic markers, amyloid PET imaging as well as accessible CSF and plasma biomarkers for AD. h Alzheimer's Disease is one of the devastating diseases of aging - projected to cause a $50 trillion Medicare deficit, requiring a massive tax increase or loss of healthcare benefits, as 75 million baby boomers cause our population to age. New drugs and high technological interventions (gene or stem cell therapy) are likely to be very expensive, associated with side-effects may be appropriate for therapy, but are unlikely to be viable approached for prevention. Based on preliminary evidence indicating success, this project aims to develop a cheap, safe and effective prevention method with over the counter dietary supplements combined with exercise to prevent Alzheimer's and to show that efficacy with our approach in people has the potential to be quickly evaluated in the clinic, using surrogate "biomarkers" for treatment efficacy.